Effect of tetracyclines on the development of vascular disease in veterans with acne or rosacea: a retrospective cohort study.

TO THE EDITOR Tetracyclines are commonly used for the treatment of acne and rosacea. In addition to their antibacterial properties, tetracyclines are increasingly being studied for other properties, such as their anti-inflammatory properties (Jackson et al., 1999; Meier, 2000; Sho et al., 2004; Griffin et al., 2005; Tessone et al., 2006; Hackmann et al., 2008; RomeroPerez et al., 2008; Griffin et al., 2010). One can hypothesize that these qualities may have secondary benefits and a protective effect on other organ systems. We sought to test the hypothesis that tetracyclines used in acne and rosacea patients may have secondary benefits on the cardiovascular system, specifically, a decreased odds ratio of developing vascular diseases. After approval by the Institutional Review Board at the Miami Veterans Affairs Health System, we used the electronic medical records from the veterans integrated service network-8 (which includes the Veterans Affairs medical centers of Bay Pines, Miami, West Palm Beach, Tampa, North Florida/South Georgia, and San Juan) to perform a retrospective observational cohort study and identify patients with the diagnosis of acne or rosacea using the International Classifications of Diseases, Ninth Revision, Clinical Modification (ICD-9) codes during the period of 1 July 2004 through 30 June 2010, allowing for at least 18 months of follow-up. We excluded any patient who had been diagnosed with vascular disease before prescription of a tetracycline, or before the diagnosis of acne/rosacea. Demographic, clinic, and pharmacy data were extracted. Vascular disease was defined and identified using the ICD-9 codes for cardiovascular disease, cerebrovascular disease, atherosclerosis, and aortic aneurysm with or without rupture/dissection. In a multiple logistic regression model, age, sex, and comorbidities (see Table 1) were included as covariates. The Hosmer–Lemeshow test was performed to assess goodness-of-fit. P-values were reported as two sided. Statistical analyses were performed using SAS software (Version 9.2, SAS Institute, Cary, NC). In total, 13,847 patients matched our inclusion and exclusion criteria (Figure 1). Patients were further subdivided—to those with prior treatment with a tetracycline, and those without tetracycline treatment (control). Demographics can be found in Table 1. There were similar race profiles between control and treatment groups within both the acne and rosacea sets (data not shown). Similar rates of comorbidities were found between study and control groups (Table 1). Of those rosacea patients who were treated with a tetracycline, 12.56% patients developed a new diagnosis of vascular disease compared with 17.15% of the control group. Overall, rosacea patients who were prescribed a tetracycline had an odds ratio of 0.69 for the development of vascular disease when compared with those not prescribed a tetracycline (odds ratio 0.69 in the univariate model, 95% confidence interval (CI) 0.61–0.79, Po0.05; odds ratio 0.78 in the multivariate model, 95% CI 0.68–0.89, Po0.05) (Table 1). Acne patients showed an odds ratio of 0.79 (95% CI 0.62–1.02) in a univariate model; however, the results were not statistically significant, and did not persist with the multivariate model. The effect of treatment duration (o3 months, 3–12 months, 412 months of daily dosing with a tetracycline) was explored; however, the group sizes were too small to make meaningful conclusions (data not shown). Cumulative dose effect was not explored, nor was daily dosage. A statistically significant decrease was found in the development of new aortic aneurysms in rosacea patients treated with doxycycline (P1⁄4 0.007) (Table 1), although the number of cases was quite small. Other vascular diagnoses were not analyzed individually. We found a potential association between the administration of tetracyclines and a decreased odds ratio for the development of vascular disease in veterans with rosacea. This study further contributes to the body of literature supporting an association between chronic low-grade inflammation and cardiovascular disease (Prodanowich et al., 2005; Wang et al., 2012). One can conjecture that the stabilization or inhibition of matrix metalloproteinases by tetracyclines has a beneficial effect on the vascular wall and/or calcifications in arteries. We did not detect any benefit of tetracyclines in acne patients, possibly due to the inherent demographics of acne patients. The average age of our acne patients may be too young to LETTERS TO THE EDITOR